Applied game-theoretic models in marketing have been popular in marketing research and are a source of interesting conclusions and explanations for interesting phenomena. In this session, I wish to share my learnings as a student of this stream of research. Using examples and participant-generated ideas, I hope to share my learnings about what makes for an interesting analytical model in marketing. In this talk, I hope to primarily generate interest in this stream of research and also discuss aspects of evaluating this kind of research.
About the Facilitator
Prof. Ramanathan S. is a faculty in the Marketing area at IIM Ahmedabad and has been associated with IIMA for the past seven years. His primary area of interest is applied game theoretic modelling in marketing. He has published in several leading INFORMS journals such as Management Science and Marketing Science.
How should an online retailer decide the set of products to be displayed for promotion, the set of products to be placed at the top of a display page, etc.? These are examples of so-called "product framing". It has been widely noted in the empirical literature that product framing (i.e., display, ranking, pricing) matters. They affect customers' attention, which in turn affect their purchasing decision. In this talk, I will present a recent work on randomized product framing and order fulfillment for e-commerce retailers, which is motivated by a clearance sale problem in a major US retailer. We analyze a relatively general setting in which customers arrive sequentially over time and the retailer's objective is to maximize his expected total profits throughout a finite selling horizon. The technical challenge of the problem comes from the fact that, in practice, the retailer's decision must satisfy a certain set of constraints including the inventory constraints (i.e., cumulative sales of a product cannot exceed its available inventory, or else there will a penalty) and the cardinality constraints (e.g., at most 20 products can be displayed for promotion at any given time). We develop a heuristic policy that can be theoretically shown to be near optimal in the setting with a large demand and large inventory. We also numerically test our heuristic policy using both synthetic and real-world data provided by a major US retailer. The results show that the proposed heuristic is very close to optimal and also outperforms many benchmarks and some state-of-the-art algorithms. One of the key take-away insights from this work is that, ideally, the framing decision must be considered jointly with the order fulfillment decision. This is important to make sure that the generated new demands across different geographical regions are in alignment with the distribution of remaining inventories across different warehouses.
About the Facilitator
Stefanus Jasin is an Associate Professor of Technology and Operations at Ross School of Business, University of Michigan, Ann Arbor. He received his Masters and PhD degrees in Statistics and Computational Mathematics from Stanford University. Stefanus' main research interests are in prescriptive and algorithmic business/market analytics, with the current focus on applications in revenue management and pricing, marketing analytic, supply chain and inventory management, e-commerce and omni-channel logistics, crowdsourced on-demand businesses, and online learning and optimization. His works in collaboration with several students and co-authors have been acknowledged by several awards, including 2019 Finalist in POMS-JD.com Best Paper Competition, 2018 Finalist in MSOM Student Paper Competition, 2018 Finalist in IBM Best Student Award Competition, 2018 Winner of INFORMS Revenue Management and Pricing Section Prize Award, 2017 Winner of INFORMS eBusiness Best Paper Award, and 2017 Second Prize in POMS-HK Student Paper Competition. Stefanus is currently serving as an Associate Editor at several journals: Management Science, Operations Research, POM Journal, and Naval Research Logistics.